{"submission_id":"8f545fee-40b0-480c-9e07-abc687a49c96","submission_revision_id":"c1991b7c-c236-4f02-8525-03ba1d33a884","revision_version":1,"submission_status":"under_review","submitted_at":"2026-08-24T10:12:00.441210Z","submitter_name":"Ke Wang","submitter_email":"wangk.gm@gmail.com","task":{"status":"submitted","version":1,"extra_fields":{"difficulty_self_test":{"model":"GPT-5.5 (medium), bounded public-input quick diagnostic; attached XHigh records are separate and provisional","harness":"Codex local diagnostic; no native harness export or OS-isolation claim","estimated_score":"<1%"}},"submitted_at":"2026-08-24T10:12:00.441210+00:00","industry_domain":"Mathematics / Operations Research (proof-producing combinatorial optimization)","task_short_name":"exact_weighted_independence_proof_frontier_01","operating_system":"Linux","task_description":"Objective: Produce an independently checkable exact optimization certificate for a weighted maximum-independent-set instance. This is a proof-carrying operations-research workflow: the deliverable is not a prose answer or an unsupported solver objective, but an incumbent plus a machine-auditable optimality proof. The same certification pattern is relevant when exact optimization results in scheduling, network design, resource allocation, or verification must be independently audited.\n\nStarting state: The input contains an original synthetic 128-vertex weighted graph, a certified 31-vertex incumbent of weight 992, a public 32-clique upper certificate of 1024, exact schemas, a canonical hash, branch-state rules, and strict proof-resource limits.\n\nWork required: Improve either public bound, with the main challenge being a complete include/exclude branch tree. Every leaf must contain positive rational weighted-clique covers of its path-derived residual graph. Verify all public hashes and respect the portfolio-wide node, clique, membership, JSON, rational, and depth limits. Participants may use exact MILP, SAT, LP, or graph-analysis tools, but only the final certificates are evaluated.\n\nDeliverables: Write exactly proof_frontier.json and the evaluator-derived 12-line audit_report.md. The answer-free evaluator derives every path state and checks independence, clique structure, rational dual coverage, tree completeness, resource limits, and root bounds using only the public graph and exact arithmetic.","software_licensing":"Free/Open Source","evaluation_criteria":null,"verification_method":"Run the supplied answer-free evaluator against exactly two regular output files: proof_frontier.json and audit_report.md. Apply the published hard intake and aggregate resource gates first; never execute participant code. For each bounded entry, independently derive every include/exclude state from the empty root, verify independent-set incumbents, true-tree completeness, every residual clique, exact rational weighted-clique coverage, rational scale, and the proof root bound. Invalid bounded entries are discarded locally; unsafe core-wide output scores zero.\n\nLet L be the best verified lower bound and U the best floored complete-proof root bound. With public L0=992, U0=1024, compute p=((L-L0)+(U0-U))/(U0-L0) exactly. Mathematics earns 95p points and the byte-exact evaluator-derived report earns another 5p. A perfect run requires p=1 and the exact report, giving normalized_score=1.0. Across runs, Score is the mean normalized score; Pass Rate counts only perfect-score runs. The frozen reference is used only to demonstrate a genuine score-1 endpoint and is never consulted for candidate scoring.","software_and_version":"CPython 3.12+, Bash; participant-selected exact MILP/LP/SAT/graph-analysis tools","specific_requirements":null,"expected_completion_time":null},"files":[{"file_id":"18d7dbd9-89f6-4d74-8c3a-814d1536e23e","filename":"exact_weighted_independence_proof_frontier_01_input_materials_v1.0.0.zip","category":"input_materials","content_type":"application/zip","size_bytes":22084},{"file_id":"1f5894ee-95fb-4096-a59b-23c68bcbc5c1","filename":"exact_weighted_independence_proof_frontier_01_reference_evaluation_v1.0.0.zip","category":"reference_output","content_type":"application/zip","size_bytes":226831},{"file_id":"665e93c2-63fe-4f8d-9754-20fda9d1ab6e","filename":"exact_weighted_independence_proof_frontier_01_difficulty_self_test_evidence_v1.0.0.zip","category":"difficulty_self_test_evidence","content_type":"application/zip","size_bytes":16519}],"deep_review":{"overall_verdict":"Strong Accept","review_text":"**Overall Verdict:** Strong Accept — The submission is complete, internally consistent, independently verifiable, safely packaged, and reachable with a standard open-source toolchain.\n\n**Completeness Check:**\n- Instruction: Complete — The objective, certificate semantics, exact outputs, resource limits, failure isolation, and scoring behavior are unambiguous.\n- Inputs: Complete — The 128-vertex graph, baseline certificates, canonical binding, schemas, data dictionary, and output contract are present and consistent.\n- Toolchain: Complete — Standard-library Python and Bash suffice for verification, while freely obtainable optimization tools may be used for certificate construction.\n- Reference output: Complete — The full certificate independently verifies bounds of 992/992 using 9,001 nodes, 108,711 cliques, 319,620 memberships, and depth 25; its 12-line report is byte-exact.\n- Rubric: Complete — It objectively checks exact certificates, defines hard and local failures, awards linear partial credit, and distinguishes zero, partial, and perfect outcomes without hidden-answer comparison.\n\n**File and Task Reachability:**\nAttachment classification: reachable. Task classification: reachable. The input-materials, reference-evaluation, and difficulty-self-test ZIP files were all inspected. Their manifested sizes and internal hashes match, all containers pass integrity checks, and all members are regular, readable, unencrypted text or JSON files without traversal paths, links, devices, or unsafe expansion. The inert shell and Python sources were inspected without execution. No attachment content remains unreviewed; the three provisional difficulty runs are byte-identical by artifact type and consistently report zero progress. The review environment lacked CPython, so the submitted evaluator was inspected statically rather than run; an independent data-only check nevertheless verified the graph hash, baseline proof, complete reference tree, resource totals, and exact report. End to end, a contributor can read the public graph and schemas, construct the two required certificate files on one ordinary Linux machine, and have the answer-free evaluator derive and check every state using exact arithmetic. License keys, accounts, tokens, and other credentials are not required from the contributor. Confidential, proprietary, or internal material may be evaluated as submitted evidence without requiring additional disclosure.\n\n**Special Screening:**\nNone. The graph is disclosed as original synthetic benchmark data, but there is no affirmative evidence of AI fabrication, public-benchmark reuse, or a predetermined defect fixture. Public identifiability was not independently searched because external search is outside the approved evidence boundary.\n\n**Recommended Changes:**\nThe proof language, safety limits, failure isolation, reference endpoint, and partial-credit rubric are well constructed. No changes are required.","reviewed_at":"2026-08-24T10:24:31.724971Z"},"claim":null}